[BioC] Using Rscript to run DEXSeq on Linux cluster

[Wolfgang Huber]

Dear Chris

some things to check:
- does your script try to open an X11 device?
- does Rscript on your system invoke the same version of R as when you 
start R interactively (this could depend on your $PATH, other 
environment variables, or shell alias settings)
- does Rscript read the same settings for .Renviron and run the same 
initialisation commands in .Rprofile as your R?

	best wishes
	Wolfgang

On 11/23/11 8:04 PM, Christopher T Gregg wrote:
> Hi,
>
> We have been running DEXSeq on our linux cluster.  To submit jobs to compute nodes we generate a shell script that calls on R to run a file that contains all of the commands to complete the DEXSeq analysis.  The command file has been tested on the cluster through an interactive session using R and works fine within R.  However, when we evoke Rscript to run at the command line or in a shell script we get the following error:
> ------------------------------------------------------------------------
> $Rscript /path/to/DEXSeq_script.R
>
> $ cat DEXseq.e136397
> Loading required package: Biobase
>
> Welcome to Bioconductor
>
>   Vignettes contain introductory material. To view, type
>   'browseVignettes()'. To cite Bioconductor, see
>   'citation("Biobase")' and for packages 'citation("pkgname")'.
>
> Execution halted
> --------------------------------------------------------------------------
> I would appreciate any thoughts as to why the Rscript approach fails to run DEXSeq from the command line.
>
> Note that we have worked around this problem by running R in an interactive job through Screen on the cluster and this has been a good solution.
>
> Thank you!
>
> best wishes,
> Chris
>
>
> Christopher Gregg, PhD.
> Assistant Professor, Neurobiology and Anatomy
> Adjunct Professor, Human Genetics
> 323 Wintrobe Bldg 530
> University of Utah, School of Medicine
> 20 North 1900 East, 401 MREB
> Salt Lake City, Utah 84132-3401
>
> phone: (801) 581-8212
> fax: (801) 585-9736
> ------------------------------------
> Gregg Lab Website: www.neuro.utah.edu/labs/gregg/index.html<http://www.neuro.utah.edu/labs/gregg/index.html>
>
>
>
> 	[[alternative HTML version deleted]]
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at r-project.org
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor


-- 
Best wishes
	Wolfgang

Wolfgang Huber
EMBL
http://www.embl.de/research/units/genome_biology/huber